kanana-1.5-8B-instruct-2505: benchmark results
Provider: Kakao.
Aggregate Rating 1465 ± 44, rank #3060 of 10916 rated models, from 21 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Polar (South Korea) | 92.45 | Total ICAT (0-100), the StereoSet idealized context associat | 94.6 |
| Polar (South Korea) - Economic Axis | 90.82 | Economic-axis ICAT (0-100), the StereoSet idealized context | 94.6 |
| Polar (South Korea) - Sociocultural Axis | 94.09 | Sociocultural-axis ICAT (0-100), the StereoSet idealized con | 94.6 |
| Polar (US) | 78.92 | Total ICAT (0-100), the StereoSet idealized context associat | 90.5 |
| Polar (US) - Sociocultural Axis | 85.44 | Sociocultural-axis ICAT (0-100), the StereoSet idealized con | 89.2 |
| Polar (US) - Economic Axis | 72.41 | Economic-axis ICAT (0-100), the StereoSet idealized context | 83.8 |
| BankMathBench (Korean) - Intermediate | 0.7 | Exact-match accuracy (%) on the 426 Korean intermediate (com | 57.5 |
| BankMathBench (Korean) - Basic | 25.8 | Exact-match accuracy (%) on the 559 Korean basic (single dep | 55 |
| BankMathBench (English) - Basic | 40.8 | Exact-match accuracy (%) on the 559 English basic (single de | 54.5 |
| TriBench-Ko - Demographic Bias | 70.5 | Macro-F1 (0-100) on the TriBench-Ko Demographic Bias risk it | 41.7 |
| BankMathBench (English) - Intermediate | 1.7 | Exact-match accuracy (%) on the 423 English intermediate (co | 40.9 |
| TriBench-Ko - Adjudicative Overreach | 61.2 | Macro-F1 (0-100) on the TriBench-Ko Adjudicative Overreach r | 33.3 |
Interactive version: theaggregate.ai/model?slug=kanana-1-5-8b-instruct-2505 · How It Works · Data refreshed daily, snapshot 2026-11-08.